1. CMCS Computer Architecture Lecture 1 Introduction and Overview January 29, CMSC411.htm
Mohamed Younis
CMCS 411, Computer Architecture

2. Lecture’s Overview Course resources, syllabus and work load
Grade structure and policy
Teaching style and philosophy
An introduction to computer architecture
The importance of studying computer architecture
Organization and anatomy of computers
The impact of microelectronics technology on computers
The evolution of the computer industry and generations
Mohamed Younis
CMCS 411, Computer Architecture

3. Instructor will stay after class to answer questions
Course Resources
Instructor: Dr. Mohamed Younis
Office: ECS 233-D
Office hours: Monday and Wednesday 4:15-5:15 PM
Research interest:
Real-time systems, Fault tolerant computing, Compiler-based analysis, Operating systems, Tool support for embedded systems
TA: Mr. Shanmugavel Ponnusamy
Office: ECS
Office hours: Tuesday and Thursday 2:45-3:45 PM
Textbook:
Computer Organization and Design, The hardware/software interface, 2nd Edition
David A. Patterson and John L. Hennessy
Morgan Kaufmann Publishers, ISBN
Web page:
Instructor will stay after class to answer questions
Mohamed Younis
CMCS 411, Computer Architecture

4. Course Syllabus 1. Performance Evaluation Measures of performance
Benchmarks and metrics
2. Instruction Set Architecture
Instruction formats & semantics
Addressing modes
3. Machine Arithmetic
ALU design
Integer multiplication & division
Floating-point arithmetic
4. Processor Design
Datapath design
Instruction exec. & sequencing
Hardwired & microcode control
Pipelining
5. Hardware Design Languages
Design with a simplified VHDL
Modeling and simulation
6. Memory Hierarchy
Cache design & evaluation
Virtual addressing
Performance evaluation
7. Input/Output
Types of I/O devices
Device access and interface
Device control
I/O performance
8. Multiprocessor (time permitting)
Interconnection networks
Programming issues
Mohamed Younis
CMCS 411, Computer Architecture

5. Course Workload Assignments Exams Projects
6 assignments will be given. The 5 best grades will be averaged and
normalized to %25 of the final grade
An average assignments requires about 3-4 hours to perform
Assignments are due in class on the due date (not later)
Exams
A midterm exam is scheduled on March 26th during scheduled class time
The final exam is scheduled on May 21st during 6:00pm - 8:00pm
Final Exam is comprehensive covering all what is covered in the class
Projects
2 projects will be given contributing 10% and 15% to the final grades
Projects are mainly hardware design assignments using a simple features
of the VHDL hardware design language
Projects must be finished and submitted on time to earn a grade
Mohamed Younis
CMCS 411, Computer Architecture

6. Grade structure and policy
Assignments are due in class (Late assignments are not accepted)
UMBC rules apply to cheating/copying
You may work together and discuss homework and the project.
You must do your own work and not copy from anyone else
You better off skipping an assignment or get a partial credit
Copying/cheating will result in a minimum punishment of a zero grade for
the assignment or project
Mohamed Younis
CMCS 411, Computer Architecture

7. Teaching Style and Philosophy
Instructor’s role
Facilitate and guide the students to the fundamental concepts
Make it simple and elaborate with examples
Relate as much as possible to available products
Prepare class notes to be as rich and comprehensive as possible
Student’s role
Focus on understanding and digesting the concept
Do not worry about the grade more than concepts, soon will be a professional
Slow down the instructor if you do not understand and raise questions
Be prepared to answer an oral quiz, when you get involved in a side talk
TA’s role
Help students with questions related to their assignments
Resolve computer and tool issues related to projects
Grade assignments and projects
Exams will question the level of understanding of fundamental concepts
Mohamed Younis
CMCS 411, Computer Architecture

8. Introduction & Motivation
Computer systems are responsible of 5-10% of the gross national product of the US
Has the transportation industry kept pace with the computer industry, a coast to coast trip would take 5 seconds and cost 50 cents
WWW, ATM, DNA mapping, … are among the applications that were economically infeasible suddenly became practical
Cashless society, anywhere computing, automated intelligent highways… are next computer science fiction expected to become a reality
Computer architecture has been at the core of such technological development and is still on a forward move
Mohamed Younis
CMCS 411, Computer Architecture

9. What is “Computer Architecture”?
Instruction Set Architecture
Machine Organization
Interfaces
Compiler/System View
“Building Architect”
Hardware Components
Logic Designer’s View
“Construction Engineer”
Mohamed Younis
CMCS 411, Computer Architecture

10. Instruction Set Architecture
... the attributes of a [computing] system as seen by the programmer, i.e. the conceptual structure and functional behavior, as distinct from the organization of the data flows and controls the logic design, and the physical implementation – Amdahl, Blaaw, and Brooks, 1964
-- Organization of Programmable Storage
-- Data Types & Data Structures:
Encoding & Representation
-- Instruction Set
-- Instruction Formats
-- Modes of Addressing and Accessing
Data Items and Instructions
-- Exceptional Conditions
SOFTWARE
The instruction set architecture distinguishes the semantics of the architecture from its detailed hardware implementation
* Slide is courtesy of Dave Patterson
Mohamed Younis
CMCS 411, Computer Architecture

11. The Instruction Set: a Critical Interface
Examples:
DEC Alpha (v1, v3)
HP PA-RISC (v1.1, v2.0)
Sun Sparc (v8, v9)
SGI MIPS (MIPS I, II, III, IV, V)
Intel (8086,80286,80386, 80486,Pentium, MMX, ...)
The instruction set can be viewed as an abstraction of the H/W that hides the details and the complexity of the H/W
instruction set
software
hardware
* Figure is courtesy of Dave Patterson
Mohamed Younis
CMCS 411, Computer Architecture

12. MIPS R3000 Instr. Set Arch. (Summary)
Instruction Categories
Load/Store
Computational
Jump and Branch
Floating Point
coprocessor
Memory Management
Special
Registers
R0 - R31 PC HI LO OP rs rt rd sa
funct
immediate
jump target
3 Instruction Formats: all 32 bits wide
* Slide is courtesy of Dave Patterson
Mohamed Younis
CMCS 411, Computer Architecture

13. Machine Organization
Capabilities & performance characteristics of principal functional units (e.g., Registers, ALU, Shifters, Logic Units, ...)
Ways in which these components are interconnected
Information flows between components
Logic and means by which such information flow is controlled
Choreography of functional units to realize the instruction set architecture
Register Transfer Level Description
Logic Designer's View
ISA Level
Functional Units
& Interconnect
Design state of art organization in 1990
* Slide is courtesy of Dave Patterson
Mohamed Younis
CMCS 411, Computer Architecture

14. Example Organization TI SuperSPARCtm TMS390Z50 in Sun SPARCstation20
Floating-point Unit
Integer Unit
Inst
Cache
Ref
MMU
Data
Store
Buffer
Bus Interface
SuperSPARC L2 $ CC
MBus Module
MBus
L64852
MBus control
M-S Adapter
SBus
DRAM
Controller
DMA
SCSI
Ethernet
STDIO
serial
kbd
mouse
audio
RTC
Boot PROM
Floppy
Cards
* Slide is courtesy of Dave Patterson
Mohamed Younis
CMCS 411, Computer Architecture

15. General Computer Organization
Processor
Computer
Control
Datapath
Memory
Devices
Input
Output
e.g., Keyboard, mouse, disk
Coordination for proper operation
e.g. Printer, Monitor, disk
Connections for Information flow
Every piece of every computer, past and present, can be placed into input,
output, memory, datapath and control
The design approach is constrained by the cost and size and capabilities
required from every component
An example design target can be 25% of cost on Processor, 25% of cost
on minimum memory size, rest on I/O devices, power supplies, and chassis
That is, any computer, no matter how primitive or advance, can be divided into five parts:
1. The input devices bring the data from the outside world into the computer.
2. These data are kept in the computer’s memory until ...
3. The datapath request and process them.
4. The operation of the datapath is controlled by the computer’s controller.
All the work done by the computer will NOT do us any good unless we can get the data back to the outside world.
5. Getting the data back to the outside world is the job of the output devices.
The most COMMON way to connect these 5 components together is to use a network of busses.
Mohamed Younis
CMCS 411, Computer Architecture

16. Levels of Representation
High Level Language
Program
temp = v[k];
v[k] = v[k+1];
v[k+1] = temp;
Compiler
lw $15, 0($2)
lw $16, 4($2)
sw $16, 0($2)
sw $15, 4($2)
Assembly Language
Program
Assembler
Machine Language
Program
Machine Interpretation
Control Signal
Specification
ALUOP[0:3] <= InstReg[9:11] & MASK
* Slide is courtesy of Dave Patterson
Mohamed Younis
CMCS 411, Computer Architecture

17. Execution Cycle Obtain instruction from program storage
Fetch
Decode
Operand
Execute
Result
Store
Next
Obtain instruction from program storage
Determine required actions and instruction size
Locate and obtain operand data
Compute result value or status
Deposit results in storage for later use
Determine successor instruction
* Slide is courtesy of Dave Patterson
Mohamed Younis
CMCS 411, Computer Architecture

18. Levels of Abstraction
S/W and H/W consists of hierarchical layers of abstraction, each hides
details of lower layers from the above layer
The instruction set arch. abstracts the H/W and S/W interface and allows
many implementation of varying cost and performance to run the same S/W
* Figure is courtesy of Dave Patterson
Mohamed Younis
CMCS 411, Computer Architecture

19. Forces on Computer Architecture
Programming languages might encourage architecture features to improve
performance and code size, e.g. Fortran and Java
Operating systems rely on the hardware to support essential features such as
semaphores and memory management
Technology always raises the bar for what could be done and changes design’s focus
Applications usually derive capabilities and constrains, e.g. embedded computing
History always provides the starting point and filter out mistakes
Computer
Architecture
Technology
Programming Languages
Operating Systems
History
Applications
History/Applications matters because
1) People buy computers to run programs
2) Most people don;’t write own programs
3) Documented IDA interface means people ship binary machine code
* Figure is courtesy of Dave Patterson
Mohamed Younis
CMCS 411, Computer Architecture

20. Technology => dramatic change
Processor
logic capacity: about 30% increase per year
clock rate: about 20% increase per year
Higher logic density gave room for instruction pipeline & cache
Memory
DRAM capacity: about 60% increase per year (4x every 3 years)
Memory speed: about 10% increase per year
Cost per bit: about 25% improvement per year
Performance optimization no longer implies smaller programs
Disk
Capacity: about 60% increase per year
Computers became lighter and more power efficient
Mohamed Younis
CMCS 411, Computer Architecture

21. Technology Impact on Processors
In ~1985 the single-chip processor and the single-board computer emerged
In the timeframe, today’s mainframes may be a single-chip computer
* Figure is courtesy of Dave Patterson
Mohamed Younis
CMCS 411, Computer Architecture

22. Processor Performance (SPEC)
RISC
introduction
Performance now improves ­ 50% per year (2x every 1.5 years)
* Slide is courtesy of Dave Patterson
Mohamed Younis
CMCS 411, Computer Architecture

23. Technology Impact on Design
DRAM capacity has been consistently quadrupled every 3 years, resulting
over 16,000 times in 20 years
Programming is more concerned with cache and no longer constrained by
RAM size
Processor organization is becoming the main focus performance optimization
Technology advances got H/W designer to focus not only performance but on
functional integration and power consumption (e.g. system on a chip)
Mohamed Younis
CMCS 411, Computer Architecture

24. Integrated Circuits: Fueling Innovation
The manufacture of a chip begins with silicon, a substance found in sand
Silicon does not conduct electricity well and thus called semiconductor
A special chemical process can transform tiny areas of silicon to either:
. Excellent conductors of electricity (like copper)
. Excellent insulator from electricity (like glass)
. Areas that can conduct or insulate under a special condition (a switch)
A transistor is simply an on/off switch controlled by electricity
Integrated circuits combines dozens of hundreds of transistors in a chip
Advances of the IC technology affect H/W and S/W design philosophy
Mohamed Younis
CMCS 411, Computer Architecture

25. Microelectronics Process
Silicon ingot are 6-12 inches in diameter and about inches long
The manufacturing process of integrated circuits is critical to the cost of a chip
Impurities in the wafer can lead to defective devices and reduces the yield
Mohamed Younis
CMCS 411, Computer Architecture

26. Computer Generations
Computers were classified into 4 generations based on
revolutions in the technology used in the development
By convention, commercial electronic computers are take to be
the first generation rather than the electromechanical machines
that preceded them
Today computer generations are not commonly referred to due
to the long standing of the VLSI technology and the lack of
revolutionary technology in sight
Mohamed Younis
CMCS 411, Computer Architecture

27. Historical Perspective
After adjusting for inflation, price/performance has improved
by about 240 million in 45 years (about 54% per year)
Mohamed Younis
CMCS 411, Computer Architecture

28. Conclusion So what's in it for you?
In-depth understanding of the inner-workings of modern
computers, their evolution, and trade-offs present at the
hardware/software boundary.
Experience with the design process in the context of a
reasonable size hardware design
Why should a programmer care?
In the 60’s and 70’s performance was constrained by the
size of memory, not an issue today
Performance optimization needs knowledge of memory
hierarchy, instruction pipeline, parallel processing, etc.
Systems’ programming is highly coupled with the
computer organization
We will take a break and talk about class philosophy.
Computer architecture is at the core of computer science & Eng.
Mohamed Younis
CMCS 411, Computer Architecture